def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
with pytest.raises(IndexError):
a[-5555] = Name.Foo
assert a == pikepdf.Array([1, Name.Foo])
a.append(4)
assert a == pikepdf.Array([1, Name.Foo, 4])
a.extend([42, 666])
assert a == pikepdf.Array([1, Name.Foo, 4, 42, 666])
def attach_notebook(pdf_in, pdf_out, notebook):
N = pikepdf.Name
main_pdf = pikepdf.open(pdf_in)
the_file = pikepdf.Stream(main_pdf, notebook["contents"])
the_file[N("/Type")] = N("/EmbeddedFile")
file_wrapper = pikepdf.Dictionary(F=the_file)
fname = notebook["file_name"]
embedded_file = pikepdf.Dictionary(Type=N("/Filespec"),
UF=fname,
F=fname,
EF=file_wrapper)
name_tree = pikepdf.Array([pikepdf.String(fname), embedded_file])
embedded_files = pikepdf.Dictionary(Names=name_tree)
names = pikepdf.Dictionary(EmbeddedFiles=embedded_files)
main_pdf.Root[N("/Names")] = names
main_pdf.save(pdf_out)
def test_jbig2_global_palette(resources):
xobj, _pdf = first_image_in(resources / 'jbig2global.pdf')
xobj.ColorSpace = pikepdf.Array(
[Name.Indexed, Name.DeviceRGB, 1, b'\x00\x00\x00\xff\xff\xff'])
pim = PdfImage(xobj)
im = pim.as_pil_image()
assert im.size == (4000, 2864)
def test_ccitt_icc(resources):
xobj, pdf = first_image_in(resources / 'sandwich.pdf')
pim = PdfImage(xobj)
assert pim.icc is None
bio = BytesIO()
output_type = pim.extract_to(stream=bio)
assert output_type == '.tif'
bio.seek(0)
assert b'GRAYXYZ' not in bio.read(1000)
bio.seek(0)
assert Image.open(bio)
icc_data = (resources / 'Gray.icc').read_bytes()
icc_stream = pdf.make_stream(icc_data)
icc_stream.N = 1
xobj.ColorSpace = pikepdf.Array([Name.ICCBased, icc_stream])
pim = PdfImage(xobj)
assert pim.icc.profile.xcolor_space == 'GRAY'
bio = BytesIO()
output_type = pim.extract_to(stream=bio)
assert output_type == '.tif'
bio.seek(0)
assert b'GRAYXYZ' in bio.read(1000)
bio.seek(0)
assert Image.open(bio)
def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
def test_jbig2_global_palette(first_image_in):
xobj, _pdf = first_image_in('jbig2global.pdf')
xobj.ColorSpace = pikepdf.Array(
[Name.Indexed, Name.DeviceRGB, 1, b'\x00\x00\x00\xff\xff\xff']
)
pim = PdfImage(xobj)
im = pim.as_pil_image()
assert im.size == (4000, 2864)
assert im.getpixel((0, 0)) == 255 # Ensure loaded
def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
with pytest.raises(IndexError):
a[-5555] = Name.Foo
assert a == pikepdf.Array([1, Name.Foo])
a.append(4)
assert a == pikepdf.Array([1, Name.Foo, 4])
a.extend([42, 666])
assert a == pikepdf.Array([1, Name.Foo, 4, 42, 666])
with pytest.raises(ValueError, match='object is not a dictionary'):
del a.ImaginaryKey
with pytest.raises(TypeError, match=r"items\(\) not available"):
a.items()
def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
with pytest.raises(IndexError):
a[-5555] = Name.Foo
def test_array_contains(self):
a = pikepdf.Array([Name.One, Name.Two])
assert Name.One in a
assert Name.Two in a
assert Name.N not in a
a = pikepdf.Array([1, 2, 3])
assert 1 in a
assert 3 in a
assert 42 not in a
with pytest.raises(TypeError):
assert 'forty two' not in a
with pytest.raises(TypeError):
assert b'forty two' not in a
assert pikepdf.String('forty two') not in a
a = pikepdf.Array(['1234', b'\x80\x81\x82'])
assert pikepdf.String('1234') in a
assert pikepdf.String(b'\x80\x81\x82') in a
def test_not_convertible():
class PurePythonObj:
def __repr__(self):
return 'PurePythonObj()'
c = PurePythonObj()
with pytest.raises(RuntimeError):
encode(c)
with pytest.raises(RuntimeError):
pikepdf.Array([1, 2, c])
d = pikepdf.Dictionary()
with pytest.raises(RuntimeError):
d.SomeKey = c
def test_filespec_types(pal, resources):
some_bytes = b'just some bytes'
some_path = resources / 'rle.pdf'
assert isinstance(some_path, Path)
fs_bytes = AttachedFileSpec(pal, some_bytes)
assert fs_bytes.get_file().read_bytes() == some_bytes
assert fs_bytes.filename == ''
fs_path = AttachedFileSpec.from_filepath(pal, some_path)
assert fs_path.get_file().read_bytes() == some_path.read_bytes()
with pytest.raises(TypeError):
fs_path.get_file(pikepdf.Array([1]))
def test_invalid_toc(resources, outdir, caplog):
pdf = pikepdf.open(resources / 'toc.pdf')
# Corrupt a TOC entry
pdf.Root.Outlines.Last.Dest = pikepdf.Array([None, 0.0, 0.1, 0.2])
pdf.save(outdir / 'test.pdf')
pdf = pikepdf.open(outdir / 'test.pdf')
remap = {}
remap[pdf.pages[0].objgen] = pdf.pages[0].objgen # Dummy remap
# Confirm we complain about the TOC and don't throw an exception
log = logging.getLogger()
_fix_toc(pdf, remap, log)
assert 'invalid table of contents entries' in caplog.text
def test_page_boxes(graph_page):
page = graph_page
assert page.mediabox == page.cropbox == page.trimbox
page.cropbox = [0, 0, page.mediabox[2] - 100, page.mediabox[3] - 100]
page.mediabox = [
page.mediabox[0] - 50,
page.mediabox[1] - 50,
page.mediabox[2] + 50,
page.mediabox[3] + 50,
]
page.trimbox = [50, 50, page.mediabox[2] - 50, page.mediabox[3] - 50]
assert page.mediabox != page.cropbox
assert page.cropbox != page.mediabox
page.cropbox = pikepdf.Array([0, 0, 50, 50])
def content_stream(self):
new_content_stream = []
for operands, operator in self.original_content_stream:
new_operands = operands[:]
if operator == pikepdf.Operator('Tj'):
new_operands = self._get_objects(operands)
elif operator == pikepdf.Operator('TJ'):
new_operands = []
for tja in operands:
if _is_array_op(tja):
objects = self._get_objects(tja)
new_operands.append(pikepdf.Array(objects))
else:
new_operands.append(tja)
new_content_stream.append((new_operands, operator))
return new_content_stream
def filter_ocg_order(self, input):
if input._type_name == 'array':
# create a copy of the array and filter the items
output = pikepdf.Array([])
for item in input:
f_item = self.filter_ocg_order(item)
if f_item is not None:
output.append(f_item)
return output
elif input._type_name == 'dictionary':
if '/Type' in input.keys():
if input.Type == pikepdf.Name.OCG:
# found the OCG entry, delete it if it's not in our keep array
if any([input.Name == oc for oc in self.keep_ocs]):
return input
else:
return None
else:
return input
def update_dest(zoom_factor,current):
dest = pikepdf.Array()
dest.append(current[0])
dest.append(pikepdf.Name("/XYZ"))
dest.append(0)
dest.append(0)
dest.append(zoom_factor)
dest_type = current[1]
if dest_type == "/XYZ":
dest[2] = current[2]
dest[3] = current[3]
elif dest_type in ("/FitH","/FitBH"):
dest[3] = current[2]
elif dest_type in ("/FitV","/FitBV"):
dest[2] = current[2]
elif dest_type == "/FitR":
dest[2] = current[2]
dest[3] = current[4]
#("/Fit","/FitB"):
return dest
def test_wrong_contains_type(self):
d = pikepdf.Dictionary()
with pytest.raises(TypeError, match="can only contain Names"):
pikepdf.Array([3]) in d
def test_nested_list2(array):
assume(isinstance(array, list))
a = pikepdf.Array(array)
assert a == array
def test_nested_list(array):
a = pikepdf.Array(array)
assert a == array
def test_issue_100(trivial):
with trivial.open_metadata() as m:
m.load_from_docinfo({'/AAPL:Example': pikepdf.Array([42])})
def test_list(array):
a = pikepdf.Array(array)
assert decode_encode(a) == a
def test_nested_list2(array):
assume(isinstance(array, list))
a = pikepdf.Array(array)
assert decode_encode(a) == a
def test_list(self, array):
a = pikepdf.Array(array)
assert a == array
def test_issue_100(trivial):
with trivial.open_metadata() as m, pytest.warns(UserWarning,
match="no XMP equivalent"):
m.load_from_docinfo({'/AAPL:Example': pikepdf.Array([42])})